# Interpreting and Disentangling Feature Components of Various Complexity from DNNs

arXiv: https://arxiv.org/abs/2006.15920 · HTML: https://mlsanity.com/paper/2006.15920/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Jun 2020.

## Built on by (most cited first, 8 of 8)

- [Eliminating Backdoor Triggers for Deep Neural Networks Using Attention Relation Graph Distillation](https://mlsanity.com/paper/2204.09975/) (2022)
- [Efficient Attribute Unlearning: Towards Selective Removal of Input Attributes from Feature Representations](https://mlsanity.com/paper/2202.13295/) (2022)
- [Revisiting Neuron Coverage for DNN Testing: A Layer-Wise and Distribution-Aware Criterion](https://mlsanity.com/paper/2112.01955/) (2021)
- [A Differential Geometric View and Explainability of GNN on Evolving Graphs](https://mlsanity.com/paper/2403.06425/) (2024)
- [Boosting RGB-D Saliency Detection by Leveraging Unlabeled RGB Images](https://mlsanity.com/paper/2201.00100/) (2022)
- [Cluster-CAM: Cluster-Weighted Visual Interpretation of CNNs' Decision in Image Classification](https://mlsanity.com/paper/2302.01642/) (2023)
- [Explaining GNN over Evolving Graphs using Information Flow](https://mlsanity.com/paper/2111.10037/) (2021)
- [Explanation-based Counterfactual Retraining(XCR): A Calibration Method for Black-box Models](https://mlsanity.com/paper/2206.11126/) (2022)

## Abstract (arXiv)

This paper aims to define, quantify, and analyze the feature complexity that is learned by a DNN. We propose a generic definition for the feature complexity. Given the feature of a certain layer in the DNN, our method disentangles feature components of different complexity orders from the feature. We further design a set of metrics to evaluate the reliability, the effectiveness, and the significance of over-fitting of these feature components. Furthermore, we successfully discover a close relationship between the feature complexity and the performance of DNNs. As a generic mathematical tool, the feature complexity and the proposed metrics can also be used to analyze the success of network compression and knowledge distillation.